HydroLAB

Drone view of a meandering reach of the River Arrow (UK) with floating tracers used for image velocimetry

New open-access paper: SSIMS-Flow image velocimetry under varying seeding densities

9 November 2026 By Salvatore Manfreda Off

Our new paper “Evaluating SSIMS-Flow velocimetry performances under varying seeding densities: a proof-of-concept field study” has just been published open access in Advances in Water Resources (2026, article 105564). The work was led by Silvano F. Dal Sasso (University of Basilicata) together with Alonso Pizarro, Robert Ljubičić and Budo Zindović (University of Belgrade), Sophie Pearce and Ian Maddock (University of Worcester) and Salvatore Manfreda (HydroLAB, University of Naples Federico II).

Read the full paper (open access, CC BY 4.0)

Why seeding matters for image velocimetry

Image velocimetry is becoming a powerful, non-contact tool for river monitoring: a camera or a drone can capture the motion of the water surface and return a complete map of surface velocities without placing instruments or operators in the stream. Its accuracy, however, depends strongly on field conditions – lighting, shadows and, above all, the density and spatial distribution of the floating tracers (“seeding”) that the algorithms track. When tracers are sparse or unevenly distributed, velocity fields can show a high variance and a systematic underestimation, especially near the riverbanks.

A controlled field experiment on the River Arrow (UK)

To quantify this effect we designed a dedicated experiment on two meandering reaches of the River Arrow (UK) under low-flow conditions. Seven drone surveys were carried out while progressively increasing the amount of artificial tracers released upstream, from natural (unseeded) to medium–high seeding densities. Conventional velocity measurements along several cross-sections provided the benchmark for the image-based estimates.

Fig. 1 – Location of the study site on the River Arrow (UK) and drone views of the upstream and downstream reaches with the velocity transects
Fig. 1 – Study site on the River Arrow (UK): location of the two reaches and drone views of the upstream (U) and downstream (D) sections with the cross-sections used as benchmark. Source: Dal Sasso et al. (2026), Advances in Water Resources, CC BY 4.0.

The videos were processed with SSIMS-Flow, the open-source optical-flow velocimetry workbench developed at the University of Belgrade, using its newly integrated seeding-analysis module, and the results were compared with those obtained with PIVlab.

Main findings

  • The MultiTMP pooling technique implemented in SSIMS-Flow achieved a high predictive accuracy, with Nash–Sutcliffe Efficiency values often above 0.8 and RMSE as low as 0.02 m/s from low to high seeding densities.
  • SSIMS-Flow provided results comparable to PIVlab’s Ensemble Correlation method, even in unseeded conditions.
  • The analysis gives practical recommendations on how to set key parameters – in particular the spatial pooling block size and the frame rate – according to the seeding conditions observed in the field.

This study continues a line of research that HydroLAB has been pursuing for several years on seeding metrics and their impact on image-velocimetry errors, and it supports the efforts of the IAHS MOXXI Working Group to make non-contact river monitoring reliable in operational settings.

How to cite

Dal Sasso, S. F., Ljubičić, R., Zindović, B., Pizarro, A., Pearce, S., Maddock, I., & Manfreda, S. (2026). Evaluating SSIMS-Flow velocimetry performances under varying seeding densities: a proof-of-concept field study. Advances in Water Resources, 105564. https://doi.org/10.1016/j.advwatres.2026.105564

Related publications

  • Ljubičić, R., Dal Sasso, S. F., & Zindović, B. (2024). SSIMS-Flow: Image velocimetry workbench for open-channel flow rate estimation. Environmental Modelling & Software. https://doi.org/10.1016/j.envsoft.2023.105938
  • Dal Sasso, S. F., Pizarro, A., Pearce, S., Maddock, I., & Manfreda, S. (2021). Increasing LSPIV performances by exploiting the seeding distribution index at different spatial scales. Journal of Hydrology, 598, 126438. https://doi.org/10.1016/j.jhydrol.2021.126438
  • Pizarro, A., Dal Sasso, S. F., Perks, M. T., & Manfreda, S. (2020). Identifying the optimal spatial distribution of tracers for optical sensing of stream surface flow. Hydrology and Earth System Sciences, 24, 5173–5185. https://doi.org/10.5194/hess-24-5173-2020
  • Pizarro, A., Dal Sasso, S. F., & Manfreda, S. (2020). Refining image-velocimetry performances for streamflow monitoring: Seeding metrics to errors minimization. Hydrological Processes, 34, 5167–5175. https://doi.org/10.1002/hyp.13919
  • Dal Sasso, S. F., Pizarro, A., & Manfreda, S. (2020). Metrics for the quantification of seeding characteristics to enhance image velocimetry performance in rivers. Remote Sensing, 12(11), 1789. https://doi.org/10.3390/rs12111789
  • Pearce, S., Ljubičić, R., Peña-Haro, S., et al. (2020). An evaluation of image velocimetry techniques under low flow conditions and high seeding densities using unmanned aerial systems. Remote Sensing, 12(2), 232. https://doi.org/10.3390/rs12020232
CategoryNews

About the author

He is Full Professor of Hydrology and Hydraulic Constructions at the University of Naples Federico II. He is currently chair of the IAHS MOXXI working group. His research primarily centers on hydrological modeling and monitoring. Recognizing the challenges posed by the complexity and limitations of traditional hydrological observations, he actively explores advanced and alternative monitoring techniques, such as the utilization of Unmanned Aerial Systems (UAS) coupled with image processing.